摘要
特征点匹配是图像匹配领域中一项重要研究内容。暴力匹配中强调缺少特征匹配不是由于太少的正确匹配,而在于很难分辨真假,故加入模板匹配进行约束,提出一种改进模板匹配算法。为了提升匹配速度和精度,该算法通过提供特征点的坐标,可以计算出每个特征点的得分值,并将其用于从输入图像中提取模板。基于输入图像和另一幅图像之间的模板匹配,使用暴力匹配算法在模板和匹配区域之间进行特征点匹配。该算法可以将高匹配数转换为高匹配质量。实验结果表明,该算法匹配速度较快、准确度较好。
Feature point matching is an important research content in the field of image matching. The emphasis on the lack of feature matching in brute force matching is not because there are too few correct matches,but because it is difficult to distinguish between true and false. Therefore,template matching is added to constrain and an improved template matching algorithm is proposed.In order to improve the matching speed and accuracy,the algorithm can calculate the score value of each feature point by providing the coordinates of the feature points and use it to extract the template from the input image. Based on template matching between the input image and another image,a brute-force matching algorithm is used to match feature points between the template and the matching area. This algorithm can convert high matching numbers into high matching quality. Experimental results show that the algorithm has a fast matching speed and good accuracy.
作者
朱鸣镝
陈婵
孙东岳
ZHU Mingdi;CHEN Chan;SUN Dongyue(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093)
出处
《计算机与数字工程》
2022年第2期377-382,共6页
Computer & Digital Engineering
基金
国家自然科学基金项目(编号:61873169)资助。
关键词
模板匹配
特征点
快速鲁棒
匹配算法
template matching
feature points
fast and robust
matching algorithm
